利用GMPLS实现光突发交换

DWDM技术虽然可以实现较大的传输容量，但是在对信息进行处理的过程中，却存在着电子瓶颈，传统网络仍然需要在交换节点处对数据进行电的交换处理，从而降低了网络的交换处理速度。为了实现在光网络上直接传送数据业务，人们提出三种IP over DWDM网络方案：波长路由交换光网络、突发交换光网络和分组交换光网络。     

全光网络的发展

实现全光网络的关键之一是用光交换代替电交换，目前自动光交换网络已经实用,了解决电子瓶颈限制问题，实现全光交换，近几年在光交换中引入了光标记交换新技术，该技术的关键在于光标记的产生、提取和识别等方面，其中有两种新方法：前置光标记法和高强度光脉冲光标记法，全光网络将向智能自动光交换网络演进。     

基于标签交换的Ad Hoc网络研究

研究基于标签交换的Ad Hoc网络，目的是提高网络的快速转发能力，减小数据包的转发时延，在功能上实现路由与交换功能的分离，本文研究通过在Ad Hoc网络中建立标签交换路径，实现数据包的标签交换，燕详细叙述了基于标签交换的Ad Hoc网络的基本架构和实现过程。     

高速网络中的交换技术

一、引言 从传统的观点看，电路交换技术不适用于数据业务网络，而分组交换技术则是当今因特网技术的主流。但是随着光传输技术的发展，带宽已不再是网络的瓶颈。技术的进步使得原本分组交换的优势和电路交换的缺陷在当今已不再有意义，而且随着应用领域的扩大，原本处于优势的分组交换技术也暴露出越来越多的问题，而有些问题若使用电路交换技术则很容易解决。光交换技术的发展，更是为我们开辟了一个新的天地。本文讨论了在高速网络中传统交换技术存在的问题，介绍了高速网络中新兴的交换技术，分析了在高速交换中如何结合分组交换和电路…     

宽带ISDN与ATM交换技术：第三讲 ATM交换技术

本文介绍了ＡＴＭ交换的基本原理，研究了ＡＴＭ交换结构的分类，并讨论了几种典型的ＡＴＭ交换网络，包括ＢＳＳ交换结构、Ｂｏｘａｎｎｅ交换单元和交换机构以及Ｂａｔｃｈｅｒ－Ｂａｎｙａｎ多级互连网络。     

光网络导通性能计算

对光波长交换网络中，存在连接源和目的节点的波长通道的概率，做了定量研究。文章首先计算单条路径、无波长交换节点的连通概率，接着计算无波长交换节点的小型网络的连通概率，然后计算掺入波长交换节点的光网络连通概率，最后对波长交换中的一些问题进行了讨论。     

光ATM交换技术异军突起

光ＡＴＭ交换是光交换网络和ＡＴＭ交换结构相结合而形成的一种颇具潜力的交换技术。本文分析了光ＡＴＭ技术产生的背景，描述基于使用超短光脉冲的广播－和－选择星形网络以及基于使用光矩阵交换的超正方体网络这两类信元交换结构，介绍了实现光ＡＴＭ交换的几项关键技术：超速信元产生、全光地址匹配和全光同步，探讨光交换技术未来的发展趋势和前景。     

诺基亚第三代移动通信系统技术解决方案系列报道：WCDMA传输网络和核心网络预规划

网络规划的第一步始于无线网络规划，其后将进入传输网络和核心网络规划，其中核心网络又分为CS电路交换域(CIRCURT CORE)和PS分组交换域(PACKET CORE)两部分。电路交换域主要是承载话音以及一些数据量小但实时性要求较高的数据业务，例如电视会议等业务，对于其它的数据业务，姗主要有PS分组交换域来承载。其整体规划流程如图1所示。     

浅议3G网络中交换网络的优化

3G网络日益临近，对于交换网络，交换网络的优化工作及交换优化人员的需求产生了巨大的冲击。文章从网络优化方面分析了3G网络相较与2G网络的不同，并结合实际的交换维护，优化经验，提出了3G网络的优化原则及方法。     

一种基于光标记交换技术的全光网络的设计方案

光标记交换是一种新的光字交换新技术，本提出一种基于光标记交换技术的全光网络的新的设计方案，叙述了这种方案的设计原理和该网络的功能。     

Commercial, legal, and international aspects of packet communications

Packet switching technology emerged rapidly in the 1970's as another viable mode of communications switching, along with circuit and message switching. Since packet switching offers economical and versatile data communication capabilities in a multiuser environment, it is particularly well suited for furnishing public data communication network services. Public packet networks are now established or being developed in most industrialized countries, and the introduction of these networks has raised policy issues relating to the structure and regulation of the national networks, and the interconnection of national networks into an international packet switching system. This paper reviews these issues and concludes that public packet switching network services will continue to be regulated in all cases; that competitive packet networks will coexist in the U.S. and in Canada, but that only one national packet network will exist in each of most other countries; that packet networks will aggravate the problem of distinguishing nonregulated data processing services from regulated data communication services; that international interconnection of public packet networks based upon CCITT standanh will occur rapidly over the next several years; and that a unified international packet switching system will eventually emerge similar to today's international telephone and telex systems.     

The universality of multidimensional switching networks

Multidimensional switching networks are networks that utilize more than one degree of freedom (e.g., space, wavelength, time, code). The main idea is to use several dimensions of practical size in hierarchical multiplexing to overcome the physical constraints present when using only one dimension of large size. We generalize the case of one-dimensional switching networks to k-dimensional switching networks. Such networks have high degree of connectivity (≫1000) and reduced complexity compared to one-dimensional networks. We introduce a technology-independent universal theory discussing two models, one of multidimensional selective switching, and the other of multidimensional broadcasting, concentrating on complexity and channel assignment. The required size of each dimension, hardware complexities, channel assignment, and corresponding routing algorithms and their time complexities are discussed. It is shown with examples bow the complexity of these networks can be reduced to a minimum by optimal allocation of the complexity in each dimension. Several realizations of three-dimensional wavelength-time-space networks using different technologies (e.g., fiber optics, acousto-optics, free-space and photonic switching) are described for both models     

Crossover minimization in directional-coupler-based photonicswitching systems

Methods of design directional-coupler-based switching networks with minimum number of crossovers are presented. Also presented is a modular construction scheme that allows large switching networks (with minimum of crossovers) to be built using identical chips containing smaller networks of the same type. The total number of crossovers and the maximum number of crossovers between an inlet-outlet pair have been analyzed in three major self-routing networks. Since many networks are based on these topologies, the results can be applied extensively to the study of many other switching networks     

On nonblocking switching networks for multichannel connections

This paper deals with switching networks for multichannel connections. The conditions under which the three-stage switching network is nonblocking for s-channel connections are given. Switching networks that are nonblocking in the strict sense as well as the switching networks nonblocking in the wide sense are considered. The conditions for two-sided and for one-sided switching networks are derived     

Fair Bandwidth Allocation in Optical Burst Switching Networks

Optical burst switching (OBS) is a promising switching technology for next-generation Internet backbone networks. One of the design challenges is how to provide fair bandwidth allocation in OBS networks; the schemes proposed for general store-and-forward IP switching networks can not be used because of the non-buffering and un-fully utilized bandwidth characteristics of OBS networks. We propose a rate fairness preemption (RFP) scheme to achieve approximately weighted max-min fair bandwidth allocation in OBS networks. We present an analysis of the burst loss probability in RFP-based OBS networks. The analysis and simulation results show that the RFP scheme provides fair bandwidth allocation in OBS networks.      

Design alternatives for <Emphasis Type="Italic">N</Emphasis>-stage WDM optical planar permutation networks

Planar permutation networks are a class of multistage switching networks with no crossover between paths that interconnect switching elements. A well-known class of planar networks is the N−Stage network that provides a good compromise between the crossbar and the Benes network. In this paper, we address the problem of designing cost-effective N-Stage optical planar networks with space-wavelength switching capability. Such networks are used for switching in communication and computing systems that employ Wavelength Division Multiplexing (WDM) technology. We investigate two classes of space-wavelength N-stage planar networks, and for each class, we design a number of switching networks and analyze their hardware complexity. In addition, we propose a new method for designing a class of space-wavelength planar networks with reduced complexity. It is shown that, for F ≤  W (where F is the total number of fibers and W that of wavelengths) the proposed method results in planar networks with an average of 67% reduction in overall cost compared to that of networks based on fixed-range wavelength converters.     

Reliable digital switching networks

A method for the design of economic and reliable PCM switching networks composed of uniform time-space elements is proposed. The described networks can be used in switching systems containing concentrators connected by an even number of PCM links to a main switching network.     

Ingénierie des systèmes de commutation optique et des réseaux locaux à large bande sur fibres optiques : analyses et perspectives

This paper first reviews how switching evolves in the ISDN environment with emphasize placed on changes in conventionnal time-division switches, then assesses techniques and technologies usable in optical switching and broadband networks : optical space-division switching systems (architecture, technologies based on dielectrics, semiconductors, photorepactive material) and optical time-division switching (optical memories based on delay lines and on bistable components multiplexing); multiple access networks (tdma, multiple access by code) ; switching using spread spectrum (bit switching); wavelength switching (multiplexer-demultiplexer, tunable laser and filters) ; packet switching in multi-wavelength networks (broadband networks standardization, local area networks with bus, passive star or multi-star configuration).     

Traffic analysis of large TST networks

The traffic performance of time-space-time (TST) switching networks built from large time-switching stages used in digital switching systems is discussed. These switching networks exhibit steep load-service characteristics and present new numerical problems for traffic performance evaluation. The precise numerical calculation of switch blocking and the amount of search effort needed to find a network path are addressed. Computational efficiency and robustness are emphasized. The approach is based on a familiar model that uses state-dependent transitions in formulating the balance equations. A comparative study with traditional methods illustrates the numerical gains to be made for switching networks of arbitrary size     

Novel type of switching network composed of digital switching matrices

A new type of one-stage switching network comprising digital switching matrices is proposed. In many cases such a network contains fewer matrices than currently known networks. The proposed one-stage networks can be used as submodules for cost-effective multistage networks.<>